Half an Hour,
Nov 06, 2023
"In what substantive ways," asked Anil Dash, "would this be different than, say, starting a WordPress blog?"
He's responding to this post from Tim Carmody, who in turn is responding to an audio message from Dave Winer. Carmody finds it remarkable that Winer is addressing what is essentially a public podcast episode to him personally. "I don't think I've ever been the personal addressee of a podcast before," he remarks, "like an @-reply, but in audio."
To me, this is in fact the core of communication and community. A large percentage of my own blog posts are in direct reply to someone. In many cases, they have written me an email asking a question and I have published my emailed response as a blog post.
There's something special about the response to one person. Kate Bowles, initiating a fascinating Mastodon thread, writes,
"What if you just share an idea with one person, whether the one sitting next to you or a passing stranger? What if the message you pick up is then only read by you, by design? What a world of reverent attention that would make.
The micropoetics of proximity (the poem between two persons, not two pages as O'Hara puts it) really interests me. What I do in my garden is seen by the very small number of people who walk down my street. My garden is their view. I look at their gardens too. We exchange ideas and sometimes seeds.
When did this stop being enough?"
It has never been enough, though. From the gathering around the campfire to the etching of marks on clay tablets to the mass production of text on a printing press to the broadcasting of ideas over radio and television, we have always sought to address more than just one person at a time: those people formally known as 'the audience'. That's why it's interesting when we combine the idea of responding to a single person with the idea of sharing that response with an audience.
The problem, as Bowles states, is "the way we use audience is as a proxy for scale not depth, it's a dream (and driver) of more audience, not more time with one" (not exactly true: there are many individuals I've dreamed of sharing more time with, but this is the most valuable thing in the world).
But it should be clear: the way this is different from blogging is that we don't think of blogging as addressed to any particular person. It's not a part of something the way a response in a personal conversation is a part of something. Or, to put the same point another way, there's no such thing as 'clicking a reply button and producing a blog post as a result'...
...except, of course, in my own newsletter, where almost every single post is in response to something specific that was created, and to this day I bemoan the fact that there was no 'in reply to' field added to the RSS specification, so that it became a publishing protocol, and not a conversation protocol. Back in the early days, I used the 'link' field to point to what I was talking about, not to my own post; virtually nobody else used RSS that way (I believe Scott Leslie did, but that's the only other I can recall).
I don't know why blogging became 'publishing', properly so-called. It wasn't for lack of effort on the part of many bloggers to keep it a conversation. We had RSS to make reading easier. We had blogrolls and blog rings. We would certainly talk back and forth with each other in our posts. We had comment feeds, before the spammers tore them apart. But the community aspect was always strained, and more and more professional writers came into the field.
And somewhere along the line, the community element of the blog became the 'share to Twitter' button, and the people who were more interested in conversation than content production migrated en masse, even if it meant they couldn't develop their ideas to any degree of detail. It didn't matter. We could now go back to replying directly to @Downes or whomever.
But back to Tim Carmody and Dave Winer. They're discussion what we would want in a next generation 'text streamer'. Winer outlines it here (I'll condense it for readability): "As a writer, these are the features I want:
- Titles are optional...
- Links...
- Simple styling, bold and italic...
- Enclosures...
- Unlimited length...
- Editable."
Winer has basically described RSS, the way he does it. What's missing? Carmody adds a few more elements (again, condensed):
- Everyday users need a default writer and reader, preferably in the same place...
- We need user and content discovery.
- We need metrics.
- We need moderation.
These features are added more from the perspective of the reader, more from the social network side of the house than from the blogging side of the house. They're the sorts of things I've seen people ask for in relation to services like Mastodon.
But both proposals suffer from a misunderstanding of the audience, a failure to grasp the importance of writing to someone, even as the two of them engage in exactly that back and forth interaction. It's easy for them to do, but a lot more difficult for the average person. You have to, in the text of your comment or post, refer back to the person you're talking to, find the link to what they're saying, and somehow embed it in your post.
For me, here, to link to the various blog posts, websites and (especially) social media posts I'm responding to has been a bit of a challenge. I would ask the reader, do you know how to find the URL of a LinkedIn post or Tweet or Mastodon Toot, let alone how to put it in your response?
That's what social media did for people. We could click on a reply button, and in the same place we were reading, address a response specifically to that person, creating a thread. A conversation. There's basically no way to do that in blogs or using RSS (at least, I've never seen one).
So, yeah, what Winer and Carmody are describing here is, as Dash suggests, no different from starting a WordPress blog. But that's a problem.
So how do we go about thinking about this?
Well, we're talking about a very specific bit of a much wider framework, which I describe in a talk here. We don't need the details of the rest of the framework; we're focused on the architecture of a personal knowledge, learning and community environment. In that talk and several others, I presented this architecture as follows:
This architecture takes into account some of the other things not even considered by Winer and Carmody, though outlined by Boris Mann in a comment here. "Modern social networks with expectations of privacy and safety" are going to need to develop systems of rights, syndication, identity and community. Things like Cosocial are attempts to implement that through organization and agreements. But we'll also, as Mann suggests, need "a toolkit for privacy and safety which is going to need public/private keys, agreement kn hash algorithms, and a host of other things."
The architecture also allows us to think about flows of content (or information, or data, or whatever you want to call it) through the architecture. For example, some publisher might produce a resource and put it in a repository, from which it is aggregated by a reader and presented to a person in the common interface. I, the reader, can then use my environment interface (by, say, clicking on a 'comment' or 'reply' button) respond to the resource, attach my identity to the response, and send it back through the environment, either as a publisher in my own right, or a contributor to a wider community.
It should make clear that there are various senses of audience that we need to be able to support:
- 'audience' as the person or resource that we are responding to
- 'audience' as the person or people that we are addressing (keeping in mind there might be multiple separate sets of people here (aka 'segments')
- 'audience' as the person or resource we are talking about
(I'm sure this sort of distinction has been made earlier and better by someone else, but I haven't looked that up.) Obviously the first and third of these are closely associated, and in most contexts can be thought of as a single item.
Why draw this out? Because it allows us to talk more specifically about one element of this architecture, the 'environment interface'. This is in essence what Winer and Carmody are talking about, and in essence what most people experience when they use a blogging or social network service. Indeed, the challenge of all this (and what makes it different from WordPress) is that we need to manage all this seamlessly through a single interface. And (if I may say) that's what social networks did better than blogs, and why they succeeded where blogs failed.
So what is this interface? There are versions of it in a variety of applications, including Element, Discord, Slack, and more. I will use as an example a screen shot from Whalebird, a Mastodon client I've been using recently:
There are seven distinct sections, which I'll label as follows:
- channel - the remote repository, network or community you are accessing
- view - the thing you're looking at from that channel (eg., replies, messages, posts, bookmarks)
- list - the contents of that view, presented (typically) in chronological order
- thread - the linkages between channel contents created by one person addressing (or 'replying to') another (in Whalebird, if you double-clock on a post, the thread column for that post appears)
- response area - the space (at the bottom of the list column) where you enter your own content, either as an original remark, or as a response to some other post
- profile - either yours, or another persons, which displays in the threads column when you click on a person's name; this provides access to the 'follow' and 'block' features
- preferences - where you can create your own profile, set screen preferences (such as dark mode or font size), proxy configuration, and other settings
The problem with WordPress and Blogger was they they consisted essentially entirely of the response area. The problem with social networks (and also with Whalebird) is that the response area is very small, very poor, and lacking privacy and security features (and in this regard, Mastodon is far ahead of traditional social media by allowing for such things as content warnings).
If we come back to the original question, what we need in the response area is not only a proper blogging tool, we need a blogging tool that is aware it is in an environment with other people and services. It should be embedded in the reader and opened via a 'reply' or 'new post' button, it should same content on our own blog but also send the response to the original post on the original channel (a la WebMentions). It should also allow addressing - that is, we can send it to whatever channel (community or network) we desire, and can support various degrees of privacy (along the lines of direct messaging to individuals or lists).
For example: this post is a response to two separate blog posts and two separate discussion threads on Mastodon. I have no way to get this back to the blog posts as comments or whatever, and to reply to the two separate Mastodon threads I had to create a post and insert this link and all the recipients by hand (essentially starting a new thread, and not adding to theirs):
To post it to LinkedIn (which I also want to do) I have to take the whole thing and copy it my hand into that platform (and all reads and discussion will be separate from everything else. Terrible!
What should happen is that my post should automatically reply to that particular audience (unless I tell it not to, which we call 'sub-tooting' or 'sub-blogging'). It should add to those respective threads, creating what is essentially a convergent discussion. And it should allow me so syndicate by publishing to whatever other services I select.
Why is all this important? Because it points to a future of social media in which:
- we don't need a separate app for each social network; we just select whatever 'channel' we want to read, and whatever channels or individuals we want to respond to
- we don't have to use the same app - we can choose Whalebird or Tusky or Feedly or whatever as we wish
- we don't need to worry about protocols - from the readers perspective there's no difference between RSS, JSONfeeds or ActivityPub
But more, it returns our experience of social media to something that is more like a conversation - a focused interaction between individuals, including somethings the audience of one - rather than a stage on which we strut and perform in hops of mass appeal (we'll leave that to the publishers, who still get what they want with this system, but at the cost of open access and open protocols).
Because, after all, we are at our best not when we're facing a blank piece of paper or computer screen, but when we're engaging in networks of similarly engaged people, coming up with new ideas together, engaging in a back and forth that takes a single idea and builds on it, allowing us as a community to create and cope with the complexities of life in the 21st century.
It's not too much to ask for this, and from where I sit, it looks like we're inching our way toward it.
Mentions
- , Nov 06, 2023, - Half an Hour, Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - Getting Started with Segments | Mailchimp, Nov 06, 2023
, - Whalebird, Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - , Nov 06, 2023
, - Different from Blogging, Nov 14, 2023
